Arachnoiditis: leveraging crowdsourcing and AI in a cross-sectional study of 1,105 cases to improve identification, understanding, and treatment
Bibliographic record
Abstract
Abstract Background Arachnoiditis, a painful and potentially disabling neurological condition, results from persistent inflammation of the spinal cord pia-arachnoid membranes following injury. While considered rare, the condition is underdiagnosed. Research on symptomatology, diagnosis, and treatments is scarce, hindering clinical management. Artificial intelligence (AI) offers promising opportunities for rare diseases, enabling large-scale pattern identification. This study used traditional research methods coupled with AI technology to characterize patient-reported clinical presentation, comorbidities, aggravating factors, and treatments for arachnoiditis. Methods This retrospective cross-sectional study utilized data from StuffThatWorks (STW), an online crowdsourcing platform for people with chronic diseases. Multiple choice and free text responses were assessed both quantitatively and qualitatively. Novel AI/machine learning algorithms were used to further analyze the data, including the STW cross-condition score (higher scores more indicative of arachnoiditis) and the STW treatment efficacy model generating effectiveness and detriment estimates, with binomial proportion 90% confidence intervals. Results Of 1250 international participants, 1105 reporting a physician-confirmed diagnosis were included. Participants were predominantly USA-based (71.4%), female (75.9%) and ≥46 years old (73.1%). Of 712 symptoms grouped into eight categories, eighteen were more indicative of arachnoiditis (by cross-condition score). The most frequent symptoms were lower back pain (43.5%), leg pain (41.6%) and back pain (39.1%). Prolonged sitting (62.5%) and prolonged standing (58.3%) were the most common aggravating factors. Comorbidities were led by degenerative disc disease (32.3%), spinal stenosis (25.3%) and fibromyalgia (25.0%). The most frequently used treatments were gabapentin (37.9%), physiotherapy (30.1%) and pregabalin (26.5%). Treatments with the highest patient-rated effectiveness (by STW model, 90% CI) were low-dose naltrexone (28.1%, CI 20.0-37.0), ketamine infusion (24.8%, CI 16.9-33.4) and fentanyl (21.1%, CI 14.7-28.1). Epidural corticosteroid injections showed the highest detriment (38.5%, CI 28.0-45.9). Conclusion As the largest observational study of arachnoiditis to date, made possible with novel methodological approaches, this work offers new insights with potential to improve diagnosis and management.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.008 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".